Investigating the Human-M. tuberculosis
interactome to identify the host targets of
ESAT-6 and other mycobacterial antigens
by Natalie Bruiners
December 2012
'LVVHUWDWLRQ presented for the degree of'RFWRURI3KLORVRSK\LQ 0edical Sciences 0HGLFDO%LRFKHPLVWU\ LQthe Faculty of 0HGLFLQH DQG+HDOWKSciences, Stellenbosch University
3URPRWRU: Prof Nicolaas Claudius Gey van Pittius Co-SURPRWRU: Prof Robin Mark Warren
ii
DECLARATION
By submitting this dissertation electronically, I declare that the entirety of the work contained therein is my own, original work, that I am the owner of the copyright thereof (unless to the extent explicitly otherwise stated) and that I have not previously in its entirety or in part submitted it for obtaining any qualification.
Date: 6 Augustus 2012
Signature: Natalie Bruiners
Copyright © 2012 Stellenbosch University All rights reserved
iii
ABSTRACT
The causative agent of human tuberculosis, Mycobacterium tuberculosis, is an intracellular pathogen that secretes virulence factors, namely ESAT-6 and CFP-10, as substrates of the ESX-1 secretion system. It is hypothesised that these substrates interact with host proteins in a targeted manner in order to elicit a required immune response, and they have been shown to be involved in processes related to pro-inflammatory responses, necrosis, apoptosis, membrane lysis and cytolysis. However, the biological function of ESX-1 substrates during host-pathogen interactions remains poorly and incompletely understood. Therefore, the present study was designed to gain insight into the role of the ESX-1 secretion system substrates in host-pathogen interactions and to identify how M. tuberculosis mediates the response of the human host.
In this study, a cDNA yeast two-hybrid library was constructed from human lung mRNA, to identify mycobacterial-host protein-protein interactions that occur within the lung alveoli. The ESX-1 secretion system substrates, ESAT-6 and CFP-10, were cloned in-frame into the pGBKT7 vector, which was used in the yeast two-hybrid system to screen the lung cDNA library in Saccharomyces cerevisiae. The ESAT-6 and CFP-10 screens identified 79 and 19 positive colonies, respectively. Of the total number of clones characterised, only two in-frame inserts were identified with the ESAT-6 screen, corresponding to the human proteins filamin A and complement component 1, q subcomponent, A chain (C1QA). In addition, the screen with CFP-10 also identified C1QA as binding partner.
Subsequent in vitro and in vivo experiments were unable to confirm the putative interactions of C1QA with ESAT-6 and CFP-10. However, the interaction between filamin A and ESAT-6 was demonstrated and confirmed by both in vivo co-localisation and co-immunoprecipitation. Furthermore, the degradation of filamin A in the presence of ESAT-6 was shown to be reflective of cytoskeleton remodelling and the induction of cell death. The work presented here suggests that as ESAT-6 gains access to the cytosol, it initiates cell death by inducing destabilisation of the cytoskeleton cell structure. This may possibly be driven by the interaction of ESAT-6 and filamin A.
Finally, we also initiated an investigation of the identified putative binding partners (filamin A and C1QA) as possible genetic markers for genetic susceptibility studies to tuberculosis. A
iv Meningitis (TBM), and 486 were controls from the South African Coloured (SAC) population within the Ravensmead-Uitsig catchment area. The results of this analysis demonstrated a novel association of a regulatory variant (rs587585) located upstream of the C1QA gene and demonstrated an increasing trend towards increased values in tuberculosis patients with the associated genotype.
This study has contributed significantly to our understanding of human-mycobacterial host-pathogen protein-protein interactions and has opened the way for future studies further exploring the consequences and function of the identified ESAT-6-filamin A interaction. It has also led to the identification of a novel genetic association with tuberculosis. Finally, it demonstrates the usefulness of the yeast two-hybrid system to identify potential protein-protein (host-pathogen) interactions that can lead to additional important and exciting research questions.
v
OPSOMMING
Die organisme wat tuberkulose veroorsaak, Mycobacterium tuberculosis, is `n intrasellulȇre patogeen wat virulensie faktore afskei, naamlik ESAT-6 en CFP-10, as substrate van die ESX-1 sekresiesisteem. Daar word vermoed dat hierdie substrate met gasheerproteïene in „n teiken wyse interaksie het om `n vereiste immuunreaksie voort te bring. Hierdie substrate is betrokke by prosesse soos pro-inflammatoriese reaksies, nekrose, apoptose, membraanlise en sitolise. Die biologiese funksie van die ESX-1 substrate tydens gasheer-patogeen interaksies word egter tans swak en onvolledig verstaan. Daarom was die huidige studie ontwerp om insig te bekom oor die rol hiervan in gasheer-patogeen interaksies en om te identifiseer hoe M.
tuberculosis die reaksie teenoor die gasheer bemiddel.
In hierdie studie was `n komplementȇre deoksiribonukleïensuur (kDNS) gis twee-hibried biblioteek gemaak vanaf long boodskapper ribonukleïensuur (bRNS) om proteïen-proteïen interaksies wat in die long plaasvind, te identifiseer. Die substrate van die ESX-1 sekresiesisteem, ESAT-6 en CFP-10, is in volgorde gekloneer in die pGBKT7 vektor en is gebruik om die long kDNS biblioteek in Saccharomyces cerevisiae te ondersoek. In die soeke na interaksies met ESAT-6 and CFP-10, was 79 en 19 positiewe kolonies onderskeidelik geïdentifiseer. Van die aantal klone, was slegs twee volgordes in-leesraam geïdentifiseer met ESAT-6. Hierdie proteïene het ooreengestem met filamin A en “complement component 1, q subcomponent, A chain” (C1QA). Bykomend hiertoe, is C1QA ook geïdentifiseer as „n bindende vennoot met CFP-10.
Daaropvolgende in vitro and in vivo eksperimente kon nie die vermeende interaksie van C1QA met ESAT-6 en CFP-10 bevestig nie. Maar die interaksie tussen filamin A en ESAT-6 kon wel gedemonstreer word deur die gebruik van lokalisering en mede-imunopresipitasie. Die afbreek van filamin A in die teenwoordigheid van ESAT-6 is ook aangetoon en blyk „n weerspieëling te wees van sitoskelet hermodellering en die induksie van seldood. Die werk wat hier aangebied word, dui daarop dat soos ESAT-6 toegang kry tot die sitosol, inisieër dit seldood deur die destabilisaisie van die sitoskelet selstruktuur. Dit word moontlik aangedryf deur die interaksie van ESAT-6 met filamin A.
vi C1QA) as moontlike genetiese merkers vir genetiese vatbaarheidsstudies vir tuberkulose uitgevoer. `n Pasiënt-kontrole studie is gedoen waarby 604 individue ingesluit is, waarvan 109 gediagnoseer is met Tuberculosis Meningitis (TBM), en die ander 486 kontrole individue was van die Suid Afrikaanse Kleurling (SAC) bevolking binne die Ravenmead-Uitsig opvanggebied. Die resultate het „n nuwe assosiasie van „n regulerende variant (rs587585) wat stroomop van die C1QA geen gelokaliseer is, getoon. Hierdie variant het `n verhoogde neiging in tuberkulose pasiënte met die geassosieërde genotipe getoon.
Hierdie studie het `n beduidende bydrae gemaak tot ons begrip van menslike-mikobakteriese gasheer-patogeen proteïen-proteïen interaksies. Hierdie resultate het die weg oopgemaak om die gevolge en funksie van die geïdentifiseerde ESAT-6-filamin A interaksie verder te ondersoek. Dit het ook aanleiding gegee tot die identifikasie van `n genetiese assosiasie met tuberkulose. Om saam te vat, hierdie werk bewys die bruikbaarheid van die gis twee-hibriede sisteem, om potensiële proteïen-proteïen interaksies te ontdek wat die moontlikheid het om aanleiding te gee tot addisionele navorsingsvrae.
vii
ACKNOWLEDGEMENTS
Lord, to You be all the glory, honour and praise.
Mama and papa, thanks you so much for all your support and words of encouragement and for always believing in me no matter what. Baie lief vir my dada en my noeksie!!!
Heinrich, my ou stouter…jou suster is baie lief vir jou. Dankie dat ek jou altyd kon bel en net kan ontlaai oor enige iets!
Richard my boeta dankie vir jou ondersteuning en die goeie tye saam op kampus.
Natasha en Graeme dankie vir al die fun times en dat julle altyd aan my snert luister. Julle twee is legends!!!
Mugeleigh dankie vir die grootste geskenk ooit, jou gebede.
My supervisor, Nico, thanks for all the support, guidance, proofreading and allowing me to run with ideas.
My co-supervisor, Rob, thanks for all you support, guidance, proofreading, fun and jokes. Eileen van Helden and Marlo Möller Karstens, thank you for your valuable input.
Craig Kinnear and Hanlie Moolman-Smook for your guidance with the yeast two-hybrid. To everyone in Lab 424, thank you for making my experience one to remember. Thank you for the support and pushing me with words of encouragement “Nearly there”, “Remember, you are?? AWESOME!!!” and “You can do this!” You are all amazing beautiful people! Olivia, Jeska, Hanlie (en jou Ouma), Jacques, Nasstasja, Neal, Francios, Nyameka and Herby thanks for all your prayers.
Suereta Fortuin, Leanie Kleynhans Cornelissen, Gaynor Gardiner, Andrea Gutschmidt, Liezl Bloem, Lynsey Isherwood, Tatiana Super, Marika “Lady” Bosman, thank you for all your support and encouragement!
The National Research Foundation, the Harry Crossley Foundation, Ernst and Ethel, Medical Research Council of South Africa, Stellenbosch Postgraduate funding and Professor Paul van Helden for financial support.
viii
TABLE OF CONTENTS
DECLARATION ii ABSTRACT iii OPSOMMING v ACKNOWLEDGEMENTS viiTABLE OF CONTENTS viii
LIST OF TABLES xv
LIST OF FIGURES xvii
LIST OF APPENDICES xxii
ABBREVIATIONS xxiii
CHAPTER 1
1. Introduction 2
1.1. The varied lifestyles of intracellular pathogens 2
1.2. Mycobacterium tuberculosis 4
1.2.1. A brief history 4
1.2.2. TB incidence 5
1.2.3. Bacteriology 6
1.2.4. The course of M. tuberculosis infection 8
1.3. Genome of Mycobacterium tuberculosis 9
1.3.1. ESX-1 secretion system 11
1.3.2. Structural characteristics of CFP-10 and ESAT-6 13
1.3.3. Role of the ESX-1 secretion system and virulence 15
1.3.4. Host binding partners of ESAT6 16
1.3.4.1. Syntenin-1 16
1.3.4.2. Toll-like receptor 2 17
1.3.4.3. Phenylalanine-rich peptides 17
1.3.4.4. Laminin 18
1.4. Mycobacterial virulence strategies inside macrophages 18
ix
1.4.2. Inhibition of phagosome-lysosome fusion and “phagosomal maturation” 20
1.4.3. Other mechanisms of survival 21
1.5. Interaction of Mycobacterium tuberculosis with the macrophage 22
1.5.1. Complement receptors 22
1.5.2. Mannose receptors 25
1.5.3. Surfactant receptors 26
1.5.4. Toll-Like receptors 28
1.5.5. Nucleotide-binding oligomerisation domain-like receptors 30
1.5.6. Other 31
1.6. Systematic approach to understanding tuberculosis 31
1.6.1. Human genetic variation and host susceptibility 32
1.6.2. Strain variation and the influence on host response 35
1.6.3. Tuberculosis and HIV co-infection 39
1.6.4. Complexity of interactome analysis 40
1.7. The present study 40
1.7.1. Problem statement 40
1.7.2. Hypothesis 41
1.7.3. Aim 41
1.7.4. Objective 42
CHAPTER 2
2 Materials and Methods 44
2.1 Strains, plasmids and cell lines 44
2.1.1 Bacterial strains 44
2.1.1.1 M. tuberculosis strain 44
2.1.1.2 Escherichia coli XL-1 blue 44
2.1.2 Saccharomyces cerevisiae strains 44
2.1.3 Plasmids 45
x
2.2.1 Electro-competent E. coli XL-1 cells 46
2.2.2 Chemically competent yeast cells 46
2.3 Polymerase Chain Reaction (PCR) 47
2.3.1 Primer design 47
2.3.2 Generation of PCR inserts 47
2.4 PCR amplification 47
2.4.1 Amplification of insert fragments 47
2.4.2 Colony PCR 48
2.4.3 cDNA amplification 48
2.5 Agarose gel electrophoresis 49
2.6 Automated DNA sequencing and analysis 49
2.6.1 DNA sequencing 49
2.6.2 Analysis of DNA fragments 50
2.7 Generation of constructs 50
2.7.1 pGemT-easy cloning 50
2.7.2 Restriction digest 51
2.7.3 Shrimp alkaline phosphatase treatment 52
2.7.4 DNA ligation 52
2.8 Plasmid transformation 52
2.8.1 Bacterial plasmid transformation 52
2.8.2 Yeast plasmid transformation 53
2.9 Plasmid purification methods 53
2.9.1 Bacterial plasmid purification 53
xi
2.10 Tissue culturing human cell lines 54
2.10.1 Culturing from frozen stocks 54
2.10.2 Splitting the cell culture 54
2.10.3 Differentiation of THP-1 cells 54
2.10.4 Generating stock cultures 55
2.11 In vivo treatments 55
2.11.1 Etoposide-induced apoptosis 55
2.12 Protein transfection 55
2.13 Evaluation of yeast strains and constructs 56
2.13.1 Phenotypic assessment of yeast strains 56
2.13.2 Testing the DNA-BD construct for transcriptional activation 56
2.13.3 Toxicity test of transformant yeast strains 57
2.14 Yeast two-hybrid analysis 57
2.14.1 Principles of the yeast two-hybrid (Y2H) technique 57
2.14.2 Library construction 58
2.14.3 Establishment of bait culture 59
2.14.4 Yeast two-hybrid assay 60
2.15 In vivo microscopy 61
2.15.1 Tissue culturing and staining for microscopy 62
2.15.2 Fluorescence microscopy and in vivo co-localisation 63
2.16 In vivo co-immunoprecipitation 63
2.16.1 Principle 63
2.16.2 Preparation of cell lysates 64
2.16.3 Membrane protein extraction 64
2.16.4 In vivo semi-endogenous co-immunoprecipitation 65
2.16.5 Western blot analysis 65
2.17 Case-Control study 66
xii
2.17.4 SNP selection 68
2.17.5 TaqMan SNP genotyping 68
2.17.6 Real Time PCR amplification 69
2.17.7 Allelic discrimination 69
2.18 ELISA assay 70
2.19 Statistical Analysis 70
CHAPTER 3
3 Yeast two-hybrid analysis to identify host binding partners of ESAT-6 and CFP-10 73
3.1 Results 73
3.1.1 Construction of lung cDNA library 73
3.1.1.1 Integrity of lung RNA 73
3.1.1.2 cDNA amplification 74
3.1.1.3 Library construction 75
3.1.2 Bait construction 76
3.1.2.1 Amplification of ESAT-6 and CFP-10 76
3.1.2.2 pGem-T-easy ligation and colony PCR 76
3.1.2.3 Construction of pGBKT7 constructs 77
3.1.3 Testing of pGBKT7 constructs 78
3.1.3.1 Testing pGBKT7 constructs for toxicity 78
3.1.3.2 Testing pGBKT7 constructs for transcriptional activation 80
3.1.4 Yeast two-hybrid screen 80
3.1.4.1 ESAT-6 80
xiii
3.2 Discussion 83
CHAPTER 4
4 Verification studies of ESAT-6 and CFP-10 binding partners 91
4.1 Results 91
4.1.1 Recombinant mycobacterial proteins 91
4.1.2 Transfection of recombinant proteins into cultured cells 92
4.1.3 Detection of endogenous proteins 94
4.1.4 In vivo co-localisation of C1QA with ESAT-6 and CFP-10 95
4.1.5 Co-immunoprecipitation of C1QA with ESAT-6 and CFP-10 96
4.1.6 In vivo co-localisation of filamin A and ESAT-6 98
4.1.7 In vivo co-immunoprecipitation of filamin A and ESAT-6 104
4.1.8 Effect of ESAT-6 on Jurkat cells and endogenous filamin A 105
4.1.9 Cytotoxicity and cell death mediated by ESAT-6 108
4.2 Discussion 109
CHAPTER 5
5 Association studies of the C1Q gene cluster 115
5.1 Results 115
5.1.1 FLNA gene 115
xiv
5.1.2 C1QA gene 117
5.1.2.1 TaqMan allelic discrimination results 118
5.1.2.2 Genotype and allele frequencies 119
5.1.3 Linkage disequilibrium and haplotype analysis 122
5.1.4 C1QA ELISA 124
5.1.4.1 Analysis of normality 124
5.1.4.2 Group and subgroup analyses 124
5.2 Discussion 126 CHAPTER 6 6 CONCLUSION 131 REFERENCES 138 APPENDICES 159
xv
LIST OF TABLES
Chapter 2
Table 2.1: Plasmids used in this study 45
Table 2.2: Primers used in this study. 48
Table 2.3: Starting product size and required concentration for sequencing reactions 50
Table 2.4: Nutritional requirements for AH109 and Y187 56
Table 2.5: Primary and secondary fluorescence labelled antibodies for use as
immunofluorescent microscopy 62
Table 2.6: Summary of fluorescent filters 63
Table 2.7: Primary and secondary antibody combinations for the detection of proteins using
Western blot analysis. 66
Table 2.8: Selected dbSNP used for this study and the TaqMan assay used to type the
particular variant. 66
Chapter 3
Table 3.1: Testing pGBKT7-ESAT-6 and pGBKT7-CFP-10 constructs for transcriptional
activation of reporter genes. 80
Chapter 5
Table 5.1: HapMap MAF of the five selected SNPs. 118
Table 5.2: Genotype and allele frequencies of the C1Q gene variants in the control,
xvi
Table 5.4: Haplotype pairs of rs587585 and rs665691. 123
Table 5.5: Results of haplotype associations. 123
Chapter 6
xvii
LIST OF FIGURES
Chapter 1
Figure 1.1: Estimated TB incidence rates, by country, in 2010. 6
Figure 1.2: Microbial characteristics of M. tuberculosis. 7
Figure 1.3: Representation of the mycobacterial cell wall 8
Figure 1.4: Schematic representation of the genomic organisation of the five ESAT-6 gene
cluster regions of M. tuberculosis. 11
Figure 1.5: Proposed model for secretion of substrates using Type VII secretion system. 12
Figure 1.6: Solution structure of the ESAT-6·CFP-10 protein complex. 14
Figure 1.7: Host cell receptors involved in M. tuberculosis infection. 24
Figure 1.8: Interplay of factors influencing tuberculosis disease. 32
Figure 1.9: Illustrates the genetic variation across a population. 33
Figure 110: Genes involved in TB susceptibility. 35
Figure 1.11: Global phylogeography of M tuberculosis. 36
Figure 1.12: Selection of MTBC isolates representative of global genetic diversity. 38
Figure 1.13: MTBC isolates and their varying ability to induce the production of
pro-inflammatory cytokines. 38
Chapter 2
xviii
Figure 2.3: Schematic flow diagram of the Y2H analyses and verification studies 60
Figure 2.4: Principle of in vivo co-immunoprecipitation. 64
Figure 2.5: Distribution of gender and age of the study cohort 67
Figure 2.6: Allelic discrimination 69
Chapter 3
Figure 3.1: Assessment of RNA integrity and purity. 73
Figure 3.2: 1.2 % TAE Agarose gel electrophoresis image of size fractionated ds cDNA
before and after purification. 74
Figure 3.3: Screening of randomly selected yeast colonies. 75
Figure 3.4: 1.5% TAE agarose gel of PCR products of ESAT-6 and CFP-10. 76
Figure 3.5: Screening of white pGem-T-ESAT-6 and pGem-T-CFP-10 colonies. 77
Figure 3.6: Construction of pGBKT7-ESAT-6 and pGBKT7-CFP-10 constructs. 78
Figure 3.7: Recombinant pGBKT7-ESAT-6 and pGBKT7-CFP-10 constructs in Y187 yeast
strains. 79
Figure 3.8: Growth rate of pGBKT7-ESAT-6 and pGBKT7-CFP-10. 79
Figure 3.9: 1.5% TAE agarose gel of PCR products of candidate clones identified during the
xix
Figure 3.10: Heterologous mating to re-test the interactions of ESAT-6. 81
Figure 3.11: 1.5% TAE agarose gel of PCR products of candidate clones identified during the
CFP-10 screen. 82
Figure 3.12: Heterologous mating to re-test the interaction of CFP-10 and C1QA of six
randomly selected clones. 83
Figure 3.13: Representation of FLNA. 86
Figure 3.14: Structure of C1Q. 87
Chapter 4
Figure 4.1: Western blot of ESAT-6 and CFP-10. 91
Figure 4.2: Western blot of Jurkat lysates transfected with recombinant ESAT-6. 92
Figure 4.3: Western blot of differentiated THP-1 lysates treated with recombinant ESAT-6
and CFP-10. 93
Figure 4.4: Cytolysis of differentiated THP-1 by recombinant ESAT-6. 94
Figure 4.5: Western blot detection of prey proteins. 95
Figure 4.6: In vivo localisation of endogenous C1QA. 96
Figure 4.7: In vivo co-immunoprecipitation of C1QA with CFP-10 and ESAT-6. 97
Figure 4.8: Non-specific background correction for ESAT-6 and filamin A. 98
Figure 4.9: In vivo co-localisation of ESAT-6 and filamin A. 99
xx
Figure 4.12: Spectral profile of ESAT-6, filamin A and the cell nuclei. 102
Figure 4.13 Inclusion bodies with ESAT-6. Figure 4.14: XYZ section through THP-1 cells
containing ESAT-6 inclusion bodies 103
Figure 4.14: XYZ section through THP-1 cells containing ESAT-6 inclusion bodies. 104
Figure 4.15: In vivo co-immunoprecipitation filamin A and ESAT-6. Figure 4.16: Expression
levels of filamin A in the presence of ESAT-6 and CFP-10. 105
Figure 4.16: Expression levels of filamin A in the presence of ESAT-6 and CFP-10. 106
Figure 4.17 Expression level of filamin A and β-tubulin in the presence of CFP-10. 107
Figure 4.18 Expression level of filamin A and β-tubulin in the presence of ESAT-6 and
etoposide. 108
Figure 4.19: Comparison of THP-1 nuclear morphology in treated and untreated with
recombinant ESAT-6. 109
Chapter 5
Figure 5.1: 1% TAE Agarose gel electrophoresis image of the PCR products of the FLNA
promoter. 115
Figure 5.2: Sequence alignment of TB cases to the reference sequence of FLNA. 116
Figure 5.3: Representative result for a TaqMan allelic discrimination plot. 119
Figure 5.4: LD structure for rs587585, rs665691, rs172378, rs12033074 and rs631090. 122
xxi
Figure 5.6: Levels of C1QA distribution between groups and subgroups. 125
Figure 5.7: Levels of C1QA distribution stratified according to subgroups and rs587585
genotypes. 125
Chapter 6
Figure 6.1: Illustration and description of the set of six Gateway-compatible yeast two-hybrid
xxii
1. Preparing buffers and media for E. coli cultures 159
2. Maps of vectors used in this study 163
3. Tissue culturing 166
4. Electro-competent E. coli 168
5. cDNA generation and library construction 172
6. Electrophoresis solutions, loading dyes and agarose gels 178
7. Promega Wizard® SV PCR, Gel and Plasmid Clean-up, High yield DNA extraction from larger volumes of bacterial culture and Yeast plasmid extraction 179
8. Bait Construction 183
9. Transform into yeast 186
10. Testing of DNA-BD Fusions 188
11. Yeast Two-Hybrid Assays 192
12. Verification and Analyses of Positive Interactions 196
13. Protocol for Immunofluorescence-Labelling of Cultured Cells 199
14. SDS polyacrylamide gel electrophoresis and Western blotting 201
15. Sandwich ELISA protocol 206
xxiii ABBREVIATIONS °C degree Celsius µF Microfarad µFd Capacitance µg microgram µL Microliters µm Micrometers µM Micromolar 3‟ 3 prime 3-AT 3-aminotrizole 5‟ 5 prime
5-FoA 5-fluoroorotic acid
A Adenine
ATPases adenosine triphosphatase
AD activation domain
Ade Adenine
ADE2 Phosphoribosylaminoimidazole carboxylase gene
AIDS acquired immune deficiency syndrome
ANOVA analysis of variance
APC antigen presenting cells
ART anti-retroviral therapy
ASC apoptosis associated spec
AUG start codon
BacA Bacitracin
BACTEC culture system to detect microbial growth from blood specimens
BAL bronchoalveolar lavage
BCG Bacille Calmette Guérin
BD binding domain
BIR baculovirus inhibitor of apoptosis repeat BLAST Basic local alignment system tool
BLASTN Basic local alignment system tool (nucleotides) BLASTP Basic local alignment system tool (proteins)
bp Basepair
BS3 suberic acid bis (3-sulfo-N-hydroxysuccinimide ester) sodium salt
C Carboxyl
C cytosine
C1q complement component 1 q
C1QA complement component 1 q chain A
C1qB complement component 1 q chain B
C1qC complement component 1 q chain C
C1r complement component 1 r
xxiv
C3bi complement protein fragment of C3
C4b cleavage product of C4
Ca2+ calcium
CARD caspase activation and recruitment domain
CCL2 chemokine(C-C motif) ligand 2
CD14 cluster of differentiation 14
CD148 receptor-type protein tyrosine phosphatase
CD209 gene encoding DC-SIGN
CD4 cluster of differentiation 4
CD63 53 kDa type III lysosomal glycoprotein
cDNA complementary deoxyribonucleic acid
CEU Utah residents with ancestry from northern and western Europe CFP10 10kDa culture filtrate protein
cfu colony forming units
CH50 total hemolytic complement
CHB Han Chinese in Beijing, China
Cl confidence interval
cm3 cubic centimeter
CO2 Carbon dioxide
CO-IP co-immunoprecipitation
CpG cytosine phosphate guanosine
CR1 complement receptor 1
CR3 complement receptor 3
CR4 complement receptor 4
CRs complement receptors
D' measure of linkage disequilibrium
dATP deoxyadenosine triphosphate
DC-SIGN dendritic cell-specific intercellular adhesion molecule-3 grabbing non- non- integrin
dbSNP single nucleotide polymorphism database
dCTP deoxycytidine triphosphate
dGTP deoxyguanosine triphosphate
dH2O distilled water
DMSO dimethyl sulfoxide
DNA Deoxyribonucleic acid
DOTS directly observed treatment and short-course
ds double stranded
dT stretch of deoxythymidine
dTTP deoxythymidine triphosphate
E. coli Escherichia coli
xxv
EccB1 gene name for locus Rv3869
EccCa1 gene name for locus Rv3870
EccCb1 gene name for locus Rv3871
EccD1 gene name for locus Rv3877
EccE1 gene name for locus Rv3882c EDTA ethylenediaminetetraacetic acid
EEA1 early endosome autoantigen 1
ELISA Enzyme-linked immunosorbent assay
ER endoplasmic reticulum
ESAT6 6kDa early secreted antigenic target ESAT-6•CFP-10 1:1 dimer complex of ESAT6 and CFP10 EspA ESX-1 secretion-associated protein A,
espACD operon encoding EspA, C, and D
EspB ESX-1 substrate protein B
EspC ESX-1 substrate protein C
espD ESX-1 substrate protein D
EspR ESX-1 substrate protein R
ESX early secretory antigenic target 6 system 1
esxA gene encoding the 6kDa early secreted antigenic target esxB gene encoding the 10kDa culture filtrate protein
EtBr Ethiumbromide
extRD1 extended region of difference 1
FcγR fragment, crystallisable gamma receptor
FLNa filamin A
FLNB filamin B
FLNC filamin C
G Guanosine
GAL4 transcriptional activator encoding galactose-metabolizing enzymes
glm general linear model
GST glutathione s-transferase
GTPase guanosine triphosphatase
H+ hydrogen isotope
H2O Water
H2O2 hydrogen peroxide
H2SO4 sulphuric acid
H37Ra attenuated Mycobacterium tuberculosis strain H37Rv virulent Mycobacterium tuberculosis strain
HA hemagglutinin
HapMap haplotype map
HCB Han Chinese in Beijing
HCI hydrochloric acid
HI-FBS heat-inactivated fetal bovine serum
xxvi
HLA human leukocyte antigen
hVPS34 class III PI 3-kinase
ID Identification
IDT Integrated DNA Technologies
IFNGR interferon gamma receptor
IFN-γ gamma interferon
IgG immunoglobulin G
IgM immunoglobulin M
IL-10 interleukin-10
IL-12 interleukin 12
IL-1β interleukin-1 beta
IL-5 interleukin-5
IL-5Rα interleukin-5 receptor alpha
in vitro latin for 'within glass'
in vivo latin for 'within the living'
IRAK4 interleukin-1 receptor-associated kinase 4 IRF interferon regulatory factors
JPT Japanese in Tokyo kb Kilobases kDa Kilodalton kV Kilovolts LAM Lipoarabinomannan LB Luria Bertani LD-PCR long distance PCR Leu Leucine
LEU2 3-isopropylmalate dehydratase
LiCl lithium chloride
LJ Lowenstein Jenson
LM Lipomannan
Ltd Limited
M Molar
mAb monoclonal antibody
MAF minor allele frequencies
ManLAM mannose-capped lipoarabinomannan
MBL mannose binding lectin
MDP muramyl dipeptide
MEL1 Alpha-galactosidase gene
μg/mL micrograms per milliliter mg/mL milligram per milliliter
xxvii
MgCl2 magnesium chloride
MHC major histocompatibilty complex
min Minute mL Milliliters mM Millimolar mm Millimeter mM Millimolar MR mannose receptor
mRNA messenger ribonucleic acid
MSMD Mendelian susceptibility to mycobacterial disease
Mtb Mycobacterium tuberculosis
MTBC Mycobacterium tuberculosis complex
MycP1 mycosin 1
MyD88 myeloid differentiation primary response gene 88
N Amino
n sample number
NaCl sodium chloride
NaPO4 sodium phosphate
NCBI National Center for Biotechnology Information
NF-κβ nuclear factor kappa beta
ng nanograms
ng/µL nanograms per microliter
NHS N-hydroxysuccinimide
NIAID National Institute of Allergy and infectious Diseases
NLR NOD like receptors
NLRP3 NLR family, pyrin domain containing 3
nM Nanomolar
NOD nucleotide binding oligomerisation domain NOS2A nitric oxide synthase 2
NRAMP1 natural resistance-associated macrophage protein 1
NZB New Zealand Black
OD600 optical density at 600nm
OmpA outer membrane protein a
OR odds ratio
P Probability
p53 tumor protein 53
PAMPs pathogen associated molecular pattern
PBS phosphate buffer saline
PCR polymerase chain reaction
PEG polyethylene glycol
pen-strep penicillin-streptomycin
pH potential of hydrogen
xxviii
PknG serine/threonine kinase G
PMA phagosomal maturation
PMA phorbol 12-myristate 13-acetate
pmol Picomol
pmol/µL pmol per microliters
PPAR-γ peroxisome proliferator-activated receptor gamma PPI protein -protein interactions
PRRs pattern recognition receptors
Pty Property
PVDF polyvinylidene difluoride
QDO quadruple drop-out
r2 coefficient of determination
Rab small GTP-binding proteins
Ras family of related small GTPases
RD region of difference
RIP2 receptor-interacting protein 2
RNA ribonucleic acid
RNI reactive nitrogen intermediates
ROI reactive oxygen intermediates
rpm revolutions per minute
RPMI 1640 Roswell Park Memorial Institute formulation 1640
RT reverse transcriptase
Rv2136c possible conserved transmembrane protein Rv3671c possible serine protease membrane protein Rv3874 6 kDa early secreted antigenic target Rv3875 10 kDa culture filtrate protein
SAC South African Coloured
SAP shrimp alkaline phosphatase
SapM 28 kDA acid phosphatase
SCINEX-P Screening for interactions between extracellular proteins
SD synthetic dropout
SD standard deviation
SDS Sequence detection systems
sec Second
SLE systemic lupus erythematous
SMART Switching Mechanism at 5' end of RNA Transcript
SNP single nucleotide polymorphisms
SOC Super Optimal broth with Catabolite repression
SOS human Son-of-sevenless gene
Sox4 SRY-related HMG-box 4 (transcription factor)
xxix
Sp-A surfactant protein A
Sp-D surfactant protein D
Sp-R120 210-kDa receptor
SV40 simian vacuolating virus 40
T Timidine
TA single 3'-overhanging thymine residue on each blunt end TACO tryptophan aspartate rich coat protein
TAE Tris acetate EDTA
TB Tuberculosis
TBM Tuberculous Meningitis
TDO triple drop-out
TE/LiAc lithium acetate
TFG-β transforming growth factor gamma
Th1 T helper cell
THP1 Human acute monocytic leukemia cell line
TLR toll-like receptor
Tm melting temperature
TMB 3, 3´, 5, 5´ - tetramethyl-benzidine
TNFα tumour necrosis factor alpha
Tris tris(hydroxymethyl)aminomethane
Trp tryptophan
TRP1 phosphoribosyl-anthranilate isomerase
UBG ultraviolet, blue, green
Ura Uracil
VDR vitamin D (1,25- dihydroxyvitamin D3) receptor
via by way of
VII 7
w/v weight/volume
WHO World Health Organisation
X-α-gal 5-bromo-4-chloro-3-indolyl alpha-D-galactopyranoside
Y2H yeast two-hybrid
YPDA yeast peptone dextrose adenine
YRI Yoruba in Ibadan
α Alpha
λ Lamda
μM Micrometers
1
Chapter 1
“We dance round in a ring and suppose, But the secret sits in the middle and knows.”
2
1 Introduction
1.1 The varied lifestyles of intracellular pathogens
Intracellular pathogens replicate either in the cytosol or in specialised phagosomal compartments to establish an infection niche. In doing so, microbes need to develop mechanisms to gain entry into host cells, as well as to escape the unfavorable effects of host immunity. Among the first lines of defense are the skin and mucosal epithelial cells of the respiratory, alimentary and urogenital tract. Once pathogens cross these physical barriers, the innate immune response is activated, to eliminate or contain infection (Kapetanovic and Cavaillon, 2007). Cellular components that mediate innate immunity includes epithelial cells, goblet cells, dendritic cells, airway macrophages, neutrophils, pneumocytes and secreted products such as antimicrobial peptides, inflammatory mediators, mucin and secretory immunoglobulin‘s (Sethi and Murphy, 2008). Of these cell types, macrophages are more permissive to intracellular pathogens due to their inherent function of engulfing foreign particles and initiating an immune response.
Macrophages, also known as professional phagocytes, regulate body homeostasis by engulfing necrotic and apoptotic cellular debris, in a silent non-inflammatory action, to aid in tissue development and maintenance. In a similar way, these cells act as alert signals for the innate immune response by recognising and engulfing invading pathogens and regulating activation of the acquired immune response (Yates et al., 2005). Beyond question, activated macrophages provide the most hostile niche for pathogen replication.
Once inside the phagosome, pathogens need to adapt to the hostile environment but it also offers most micro-organisms the ability to escape humoral recognition by circulating antibodies and complement. Some pathogens reside in the phagosome and maintain features of either early-endosomes or late-early-endosomes by diverting the endocytic pathway, normally targeted by lysosomal compartments (Cossart and Roy, 2010). A few pathogens, like Brucella abortus and
Legionella pneumophila, avoid lysosomal vesicles by using an alternative route through the
endoplasmic reticulum (ER) to disguise the phagosomal membrane as an ER membrane (Kumar and Valdivia, 2009).
3
A selected number of bacteria succeeds in escaping from the phagosome and replicate within the cytosol. This is true for Listeria monocytogenes, Shigella flexneri, Rickettsia conorii and
Francisella tularensis. Once entry into the cytosol has been established, pathogens need to
circumvent autophagy. Autophagy is a conserved event in which double layers of cellular membrane encapsulate unwanted cytosolic particles and degrade them by fusion with lysosomal vesicles (Cossart and Roy, 2010). In escaping autophagy, pathogens rapidly replicate, and employ actin polymerisation to establish intra- and inter-cellular distribution. Actin motility is employed by Listeria and Shigella, which hijack host protein complexes to produce actin comet tails for escape. These host protein complexes, which are considered uniquely for actin motility, have also been connected to autophagy recognition (Mostowy and Cossart, 2011).
However, pathogens within the phagosome are challenged by phagolysosome fusion, which is designated to dispose of microbial invaders. Several intracellular pathogens like Salmonella,
Chlamydia and Mycobacterium have developed various strategies for inhibiting phagolysosome
fusion (Duclos and Desjardins, 2000; Deretic et al., 2006). Whether pathogens truly inhibit phagolysosome fusion, is open for debate, since authors suggest that it is rather a continual communication with the endocytic pathway delaying the event (Drecktrah et al., 2007; Pryor and Raines, 2010). However, it is clear that proteins and lipids involved in the regulation of lysosomal trafficking are targeted by secretory products of the pathogen (Pryor and Raines, 2010). Mycobacteria seem to be more proficient in this survival mechanism (Nguyen and Pieters, 2005).
Macrophages are long-lived compared to neutrophils (which live for a much shorter lifespan and are also professional phagocytes) (Allen, 2003). Therefore, inhabiting macrophages provides a much more stable timeframe for intracellular replication. In the final step of replication and spread, pathogens need to be released from their replication niche in order to transmit to other cells. Examples of cell-to-cell spread have been observed with Listeria, malaria plasmodia and
Toxoplasma gondii (Mounier et al., 1990; Sibley, 2004). Malaria and T. gondii strictly require
being within host cells compared to Listeria, Mycobacterium tuberculosis, Brucella spp. and
Salmonella enterica, which have developed the capacity to exist in the extracellular environment
prior to entering new host cells. Microbes like these can even live in an abiotic milieu for longer periods of time. Despite the fact that M. tuberculosis is ultimately seen as an intracellular
4
bacterium, it flourishes in the extracellular debris of granulomas, reaching up to 1013 bacilli (Kaufmann, 2011) and can persist asymptomatically in the human host for years.
1.2 Mycobacterium tuberculosis
1.2.1 A brief history
Robert Koch (1843-1910) had variable research interests and studied multiple organisms responsible for anthrax, plague, cholera, malaria, sleeping sickness and numerous animal diseases such as rinderpest and surra of cattle and horses. Regardless of this wide range of interests, his name is usually associated with the development of staining and culture methods for bacteria and for the isolation of the slow-growing agent of tuberculosis – Mycobacterium tuberculosis (Taylor et al., 2003). His work on M. tuberculosis was presented to the Physiological Society of Berlin in March 1882 (Ligon, 2002).
A few years later, in 1895, Wilhelm Konrad Röntgen discovered X-rays and together these two scientific achievements were applied to clinical medicine. From 1905, doctors could start making a precise diagnosis by demonstrating abnormalities in a patient‘s chest radiograph and isolating tubercle bacilli from sputum (Murray, 2004). During the 1880s, Louis Pasteur developed a method to intentionally attenuate the virulence of a living microbe to generate a successful vaccine; the first was against fowl cholera and later against rabies and anthrax (Pasteur et al., 2002). In 1908, Albert Calmette and Camile Guérin applied Pasteur‘s technique to generate a vaccine against tuberculosis. By a fortunate discovery, they learned that growth in ox bile diminished virulence of M. bovis and by performing 230 serial passages they isolated a single colony unable to cause fatal tuberculosis in a number of animals – among these were guinea pigs, rabbits, cows, horses, monkeys and chimpanzees (Dubos, 1986). The vaccine was called Bacille Bilié Calmette et Guérin, shortened to Bacille Calmette Guérin (BCG). In 1921, BCG vaccination was given to a number of babies and young children and by 1924 the vaccine demonstrated apparent protection with fewer side effects. It has been commonly observed that BCG immunisation is helpful in infants, providing protection against severe forms of tuberculosis, particularly miliary and meningeal disease (Dubos, 1986; Murray, 2004).
5
Much hope was placed on mass vaccination; but this was quickly shattered by the lack of BCG‘s protective effects. There has been no consensus about the usefulness of BCG vaccination, as BCG-induced immunity does not prevent subsequent development of infection; it only retards the spread (Gomes et al., 2004). Anti-tuberculous drugs was introduced in the 1940s (streptomycin) and 1950s (isoniazid), these and several other antibiotics were used as a clinical regimen (Lönnroth et al., 2009). Today, many countries document increases in multidrug resistant strains and in some cases now resistance to all available drugs (Velayati et al., 2009). Furthermore, no new drugs have been licensed since the identification of ethambutol in the 1960s (Comas and Gagneux, 2011). With the increasing prevalence of drug resistance and TB-human immunodeficiency virus (HIV) co-infections, there is an urgent need for new control strategies, improved diagnostics, effective drugs, shorter treatments and better vaccines (Kaufmann and Parida, 2007).
1.2.2 TB incidence
The World Health Organisation (WHO) – Global Tuberculosis Control 2011 Report – estimated that during 2010 there were an estimated 8.8 million incident cases (range, 8.5 – 9.2 million). The greatest proportion of incident cases occurred in Asia (59%) and Africa (26%), with smaller rates occurring in regions of the Eastern Mediterranean (7%), Europe (5%) and America (3%) (Figure 1.1). The five countries that had the largest TB incidence in 2010 were India (range, 2.0-2.5 million), China (range, 0.9-1.2 million), South Africa (range, 0.4-0.59 million), Indonesia (range, 0.35-0.52 million) and Pakistan (range, 0.33-0.48 million). Of the reported 8.8 million cases in 2010, the best fit estimate of people living with TB-HIV co-infection was 1.1 million (13%). Strikingly, approximately 82% of the HIV-positive individuals were from the African Region (Global Tuberculosis Control 2011 Report).
6
Figure 1.1: Estimated TB incidence rates, by country, in 2010 (WHO report 2011).
Global TB control programs primarily focus on understanding transmission through early detection and effective treatment. Although control measures such as BCG vaccination and the WHO Directly Observed Treatment Short-Course (DOTS) strategy have been implemented, the rise in HIV transmission and the emergence of drug resistant phenotypes are still troublesome. However, since the addition of DOTS into the Stop TB strategy, the approach led to the treatment of 36 million patients between 1995 and 2008 and prevention of up to 8 million deaths (Ahmad, 2011). A combination of disease surveillance and computational modeling predicts that the 1990 targets of prevalence and mortality could be halved by 2015. The probability is, that if these targets are reached and maintained, the rate of disease will decrease, which will require a set of effective interventions that are both cost-effective and capable of producing effects on larger scale (Dye and Floyd, 2006). In 2005, the African Health Ministers at the 55th Session of the WHO Regional Committee for Africa in Maputo declared tuberculosis as an ―Emergency‖, ensuring immediate and determined efforts to combat the disease (Chaisson and Martinson, 2008).
1.2.3 Bacteriology
Tuberculosis is caused by a number of closely related Gram-positive bacilli in the genus of Mycobacterium, known as the Mycobacterium tuberculosis complex (MTBC). The MTBC
7
consists of a number of species and subspecies that include M. tuberculosis, M. africanum, M.
bovis, M. canetti, M. caprae, M. microti, M. pinnipedi and recently discovered M. mungi
(Alexander et al., 2010).
M. tuberculosis is a non-motile rod shaped bacterium (Figure 1.2A). The rods are 2-4
micrometers (µm) in length and 0.2-0.5 µm in width. M. tuberculosis is an obligate aerobe. For this reason, it is classically found within well-aerated upper lobes of the lungs, where it functions as an intracellular facultative pathogen. The tubercle bacilli has extremely slow doubling times of ~20 hours (Todar, 2011), compared to the closely related M. marinum that doubles every ~ 4 hours (Ramakrishnan and Falkow, 1994). As a result, M. tuberculosis can establish its niche within the host before recognition by host immunity.
Figure 1.2: Microbial characteristics of M. tuberculosis – (A) Acid-fast staining of rod shaped M. tuberculosis
recovered from glass culture tubes (Taylor et al., 2003). (B) Colonies of M. tuberculosis are rough, waxy, thick, wrinkled, have an irregular margin, and are faintly buff-coloured (Ojha et al., 2008).
Traditionally, M. tuberculosis is grown on Lowenstein Jensen (LJ) culture media requiring several weeks before the bacilli can be isolated, resulting in delay of treatment. A faster result is nowadays obtained using Middlebrook medium or the BACTEC™ MGIT™ 960 System. Colonies of M. tuberculosis have an unusual, waxy coat on the cell surface that primarily consists of mycolic acids (Figure 1.2B) (Todar, 2011). This distinct feature highlights the large number of enzymes involved in lipid metabolism (Guo et al., 2010). It is also the dense lipid cell wall that makes the cells impermeable to gram staining, hence the use of acid fast staining techniques for identification (Figure 1.2A) (Misawa, 1952).
The majority of the cell wall components consist of peptidoglycans and lipids. The lipid component primarily consists of hydrophobic mycolic acids, which are α-branched lipids and
8
makes up 50% of the dry weight of the mycobacterium (Figure 1.3). As a result, mycobacteria are slow growers due to the impaired entry of nutrients through the thick layer of mycolic acids, but it also provides resistance to cellular degradation processes within the human host (Brennan and Nikaido, 1995). Mycolic acids are dispersed mostly at the external portions of the cell wall, while the internal portions mostly consist of arabinogalactan, phosphatidyl-myo-inositol mannosides (PIMs) and peptidoglycans. Adjacent to the mycolic acids are other composites such as mannose-capped lipoarabinomannan (ManLAM), the related lipomannan (LM) and mannosylated glycoproteins. The outer capsule and surface of the bacilli presents with mannan and arabinomannan caps (Torrelles and Schlesinger, 2010).
Figure 1.3: Representation of the mycobacterial cell wall. The cell wall consists primarily of a large cell-wall
complex that comprises of three covalently linked structures i.e. mycolic acids (green), arabinogalactan (blue) and peptidoglycan (grey). The mycolic acids are covalently linked and results in a hydrophobic layer of low fluidity, referred to as the mycomembrane (Abdallah et al., 2007).
1.2.4 The course of M. tuberculosis infection
The first stage of tuberculosis begins with the inhalation of M. tuberculosis containing aerosols into the respiratory tract. Once in the lung, the bacilli are phagocytized by alveolar macrophages resident in the lung airways (Cooper, 2009). During this stage, the destruction of the
9
mycobacteria depends on the intrinsic bactericidal capacity of alveolar phagocytes and the virulence spectrum of the inhaled M. tuberculosis strain. During the initial infection, the bacillus multiplies in the lung, causing limited inflammation. Infection is established by employing multiple survival strategies within macrophages (Fatty et al., 2003), as well as the initial interactions with host cells. These survival mechanisms include triggering anti-inflammatory responses, blocking reactive oxygen and nitrogen production, and reducing acidification of M.
tuberculosis-containing phagosomes (Flynn and Chan, 2001; Cooper, 2009).
During the persistent phase of infection, the bacilli escape bactericidal properties and multiply, resulting in the destruction of alveolar macrophages. This prompts a localised inflammatory response that results in the supply of neighbouring mononuclear cells from blood vessels, providing new host cells for the bacterial growth (Russell et al., 2010). The monocytes mature into either antigen presenting cells, alveolar macrophages or dendritic cells. Alveolar macrophages ingest, but do not kill the bacteria, allowing them to effectively grow with limited tissue damage. T lymphocytes are activated and recruited 2-3 weeks post infection via antigen presenting dendritic cells travelling from the site of infection (Kleinnijenhuis et al., 2011).. The recruited T cells form the building blocks of the early granuloma, where macrophages become activated to kill intracellular M. tuberculosis (Ulrichs and Kaufmann, 2006; Sasindran and Torrelles, 2011). Continued activation of T cells leads to the formation of granulomas, which are the pathological signature of this disease, and subsequently to the emergence of cell-mediated immunity. This marks the latent stage of infection, where growth and spread to additional tissue sites are limited. The third and final stage is when latent and controlled infection is activated. The two main reasons for activation of disease is marked by (1) a decline in the host‘s immunity due to genetic, environmental or biological causes and (2) failure to maintain and develop immune signals, resulting in lung cavitation and development of disease (Sasindran and Torrelles, 2011).
1.3 Genome of Mycobacterium tuberculosis
Tuberculosis research has made great advances since the publication of the M. tuberculosis H37Rv genome in 1998 (Cole et al., 1998). Since then, sequences from M. bovis, M. bovis BCG,
10
the process of completion. The genome of the laboratory strain, H37Rv, was revealed to possess a sequence length of 4, 411, 529 bp (Cole et al., 1998). Comparative genomics have revealed a number of polymorphisms across various members of the MTBC that adds to the phenotypic diversity observed across the various strains. Larger polymorphic regions, those greater than 0.5 kb, have also been identified between strains and are termed regions of difference (RD loci).
Mahairas and colleagues applied subtractive hybridisation to identify regions of difference that accounted for the avirulent phenotype observed in M. bovis BCG (Mahairas et al., 1996). Their studies showed three regions of difference in the genome of H37Rv that were absent from M.
bovis BCG. Of these regions, RD3 corresponded to one (PhiRv1) of the two prophages elements
(PhiRv1 and PhiRv2), and were found to be varied among M. tuberculosis clinical and laboratory strains. RD2 was only deleted from isolates of M. bovis BCG that was re-cultured from 1925. Lastly, and most importantly, RD1 was found to be deleted from all M. bovis BCG strains and were only present in the ancestral pathogenic strains, leading to the hypothesis that this deletion caused the attenuation of BCG. However, complementation assays with RD1 into modern BCG strains did not reconstitute the full virulent phenotype of M. bovis BCG (Mahairas et al., 1996).
With the completion of the genome, other differences were found in BCG-duplicated regions (Brosch et al., 2000), and lineage specific deletions and point mutations (Behr et al., 1999) that all appeared to contribute to the attenuation of BCG (Brosch et al., 2007). However, many mechanisms are specific to M. tuberculosis, one of which is dependent on a fully functional RD1 locus together with an effective two-component regulatory system (Frigui et al., 2008). The RD1 locus is situated in a cluster of genes that are part of a specialised secretion system named the ESX-1 secretion system (Hsu et al., 2003; Pym et al., 2003; Stanley et al., 2003; Brodin et al., 2004; Guinn et al., 2004). The individual genes encoded by this locus have been shown by several groups to participate in virulence (Hsu et al., 2003; Pym et al., 2003; Sassetti and Rubin, 2003; Stanley et al., 2003; Guinn et al., 2004).
11
Figure 1.4: Schematic representation of the genomic organisation of the five ESAT-6 gene cluster regions of M.
tuberculosis (ESX-1 to - 5) (Gey van Pittius et al., 2001).
1.3.1 ESX-1 secretion system
M. tuberculosis has five specialised ESX secretion systems (ESX-1 to ESX-5, Figure 1.4), which
have been dubbed Type VII secretion systems (Abdallah et al., 2007). The ESX systems are named after the first known secretion product of the ESX systems, the 6kDa early secreted antigenic target (ESAT-6). The trademark of these systems is that they secrete products with homology to ESAT-6 (Feltcher et al., 2010).
The proteins associated with the ESX-1 secretion system can broadly be classified into three groups, namely – (1) secreted, (2) regulatory and (3) structural proteins. Combining computational results with those known from literature, Das and colleagues proposed a model for secretion of proteins through the ESX-1 secretion machinery. The model is depicted in Figure 1.5 (Das et al., 2011).
12
Figure 1.5: Proposed model for secretion of substrates (CFP-10, ESAT-6, EspA, EspC, EspB) using Type VII
secretion system. The various steps involved are indicated by numbers and are described below: (1) Initial activation
by WhiB6; (2) Transcriptional regulation of the secretory proteins by Rv3866 and PhoP; (3) Multimeric assembly of secretory proteins by Rv3868; (4) Pairing up of secretory protein EspB with Rv3879c; (5a) Recruitment of CFP-10/ESAT-6/EspA/EspC multimer by Rv3870 and Rv3871 to the inner membrane pore (Rv3877); (5b) Recruitment of EspB/Rv3879c pair by Rv3870 and Rv3871 to the inner membrane pore (Rv3877); (6) Interaction of MycP1 with Rv3877; (7) Inner membrane translocation of secretory proteins; (8) Cleavage of EspB by MycP1; (9) Mycomembrane translocation of secretory proteins. The dashed arrows show interaction between the components connected by the arrow (Das et al., 2011).
The ESX-1 system functions to secrete effector proteins, including the 10kDa culture filtrate protein (CFP-10 also known as esxB) directly into the host phagosome or cytosol (Stanley et al., 2007; Lewinsohn et al., 2006). In addition to ESAT-6 and CFP-10, four other substrates are secreted by the ESX-1 system: EspR, EspA, EspB and EspC (Gao et al., 2004; Fortune et al., 2005; MacGurn et al., 2005; McLaughlin et al., 2007; Xu et al., 2007). EspR is a transcriptional activator that regulates the secretion of the system and are found to influence the expression of the espACD loci, which in turn regulates export of the ESAT-6·CFP-10 complex. The secretion of EspR is dependent on activation of the ESX-1 system that results in reduced expression of
espACD genes (Blasco et al., 2011). So, when the pathway is inactive, the cytosolic
13
Additionally, EspA secretion is dependent on the secretion of the ESAT-6·CFP-10 complex and
vice versa (Fortune et al., 2005).
The ESX-1 locus contains two cytoplasmic AAA ATPases, EccA1 and EccCb1, that probably supply energy for the secretion process and are involved in targeting products for secretion (Stanley et al., 2003; Luthra et al., 2008). Eccb1 binds the C-terminal signal peptide of CFP-10 and EccA1 binds the C-terminal region of EspC, these binding abilities are required for secretion of both the ESAT-6·CFP-10 complex and EspC (Champion et al., 2006; DiGiuseppe Champion et al., 2009).
EccD1 is a multi-transmembrane protein that is proposed to be involved in the formation of the transmembrane pore for the translocation of mycobacterial proteins and virulence factors (Brodin et al., 2005). The integral membrane protein, EccCa1, interacts with ATPase EccCb1 (Stanley et al., 2003). EccB1 and EccE1 are predicted to be transmembrane proteins located in the periplasm of the cell wall (Krogh et al., 2001). EspD has been shown to interact with EccE1 and is required for the secretion of the ESAT-6·CFP-10 complex (MacGurn et al., 2005). Mycosin-1 (MycP1) is a transmembrane protein and is located in the cytoplasmic membrane of the cell wall. MycP1 is a serine protease that has been shown to cleave EspB after translocation and has been shown to be essential for ESX-1 secretion (Ohol et al., 2010).
The ESX-1 secretion apparatus is a unique system, in that secretion of substrates is mutually dependent on all substrates of the system (Fortune et al., 2005; Abdallah et al., 2007) and deletion of any part of the system causes attenuation of the organism (Hsu et al., 2003; Stanley et al., 2003; Guinn et al., 2004; Wards et al., 2000).
1.3.2 Structural characteristics of CFP-10 and ESAT-6
The open reading frames encoding for ESAT-6 and CFP-10 lie in an operon and are co-transcribed. The protein products form a tight 1:1 complex (Renshaw et al., 2002; Lightbody et al., 2004). Renshaw and colleagues determined the structure of the ESAT-6·CFP-10 complex (Figure 1.6) and revealed that the inner core had a well-defined complex of two similar
14
turn-helix hairpin structures for each individual protein, which lie parallel to each other, forming a four helix bundle. The contact surface between CFP-10 and ESAT-6 is hydrophobic in nature (Renshaw et al., 2005) and the complex formation stabilises the protein, which is important for secretion, as CFP-10 contains a C-terminal signal peptide that targets the whole complex for secretion.
Figure 1.6: Solution structure of the ESAT-6·CFP-10 protein complex. (A) A best-fit superposition of the protein
backbone, with CFP-10 shown in red and ESAT-6 in blue. (B) A ribbon representation of the backbone topology of the ESAT-6·CFP-10 complex based on the converged structure closest to the mean, which illustrates the two helix– turn–helix hairpin structures formed by the individual proteins (Renshaw et al., 2005).
It has been demonstrated that once ESAT-6 and CFP-10 associate as a complex; it is the C-terminus of CFP-10 that is responsible for the binding to host cells. This was illustrated by truncation of CFP-10 (residues 1-86) bound to full length ESAT-6 and full length CFP-10 bound to ESAT-6 with residues 1-84 being deleted. The deletion of CFP-10 residues impaired the binding of the complex to macrophage host cells, while the deletion of ESAT-6 showed no distinguishable difference when compared to the wild-type complex (Renshaw et al., 2005). Thermodynamic analyses revealed that binding between ESAT-6 and CFP-10 can take place below any melting temperature (Tm)of the complex, which is 53.4°C. An interesting observation
was the loss of binding capacity to phospholipid membranes upon complex formation. Moreover, ESAT-6 adopted a more helical structure and enhanced thermal stability compare to CFP-10 or the complex, in the presence of phospholipid membranes. The processes that govern this observation and whether these features of ESAT-6 are observed on the surface of macrophages remain unknown. With respect to biochemical stability, the complex was found to be more
15
resistant to proteolytic cleavage by trypsin. The work presented by Meher et al., demonstrates overall structural change, enhanced thermodynamic and biochemical stability upon complex formation and highlight features involved in the physiological role of ESAT-6, CFP-10 and/or the complex in host cells (Meher et al., 2006).
1.3.3 Role of the ESX-1 secretion system and virulence
The 1 secretion system is a major virulence determinant of M. tuberculosis. Described ESX-1 effects are related to suppression of pro-inflammatory responses, necrosis, apoptosis, membrane lysis and cytolysis (Simeone et al., 2009). It has long been thought that ESAT-6 and CFP-10 are the effector molecules of the system, but additional proteins are also secreted and may serve as effector substrates of the ESX-1 machinery.
The most documented function of ESAT-6 in literature is linked to lysis of cell membranes. It has been demonstrated that deletion mutants of ESAT-6 did indeed multiply in macrophages but were unable to spread to uninfected macrophages (Guinn et al., 2004). In addition, deletion mutants displayed reduced invasiveness due to the lack of cytolytic ability. The interaction of ESAT-6 and CFP-10 is dependent on pH and conversion to a more acidic environment leads to dissociation of the complex (de Jonge et al., 2007). Moreover, ESAT-6 has a greater ability to disrupt and lyse liposomes, whereas CFP-10 does not. Therefore, it is thought that the biological function of ESAT-6 is dependent on the acid environment within the phagosome (de Jonge et al., 2007). In contrast, Lightbody et al., demonstrated that the complex between ESAT-6 and CFP-10 showed no decrease in stability and appeared to be more stable at lower pH (Lightbody et al., 2008). These observed inconsistencies could be explained by the fact that Guinn et al., used acetylated ESAT-6 from culture filtrates (Guinn et al., 2004), whereas the latter study used non-acetylated ESAT-6 expressed from Escherichia coli (Lightbody et al., 2008). Furthermore, blot overlay assays demonstrated that CFP-10 primarily interacted with non-acetylated ESAT-6, with less binding affinity to acetylated ESAT-6 (Okkels et al., 2004).
Literature suggests that M. tuberculosis remains in the phagosome, inhibiting maturation and phagosomal acidification, but escape from the phagosome remains to be elucidated. Lee and
16
colleagues showed that the metabolic status of M. tuberculosis correlates with the degree of phagosomal maturation. They demonstrated that compartments occupying mycobacteria differ markedly from fully matured phagolysosomes. They showed that metabolically active M.
tuberculosis is less fusiogenic with lysosomes than metabolically inactive M. tuberculosis. In
addition, they also observed that 5 days post infection, 25% of M. tuberculosis was found in the cytosol of macrophages without phagosomal membranes (Lee et al., 2008). This confirms the finding of van der Wel and coworkers who proposed translocation of M. tuberculosis from the phagosome to the cytosol (van der Wel et al., 2007).
In summary, the ESX-1 secretion system is clearly involved in pathogenicity and host-cell interactions, but there are still many details outstanding on the biology of these secretion systems and the function of the secreted products thereof. The elucidation thereof will have important consequences for the development of vaccines and therapeutic agents.
1.3.4 Host binding partners of ESAT-6
ESX-1 mediated secretion occurs early during infection before uptake into phagocytes, therefore secreted products can engage host receptors of alveolar macrophages and dendritic cells, as well as host components of the phagosome and cytosol. These interactions take place between secretory products of M. tuberculosis and the host cells at the primary site of infection; in order for infection to be established. In the following sections we discuss findings of ESAT-6 interactions with host proteins and the role of these interactions in the host-pathogen relationship.
1.3.4.1 Syntenin-1
Syntenin-1 is a host protein which functions in multiple roles related to vesicular trafficking, cytoskeletal dynamics, protein turnover, cell adhesion and signaling pathways involved in cell differentiation (Bernfield et al., 1999; Lander and Selleck, 2000; Yoneda and Couchman, 2003; Sarkar et al., 2004). Syntenin-1 was reported to interact with CD148, a transmembrane protein tyrosine phosphatase (Harrod and Justement, 2002). Numerous investigators observed negative regulation of T cell activation by CD148, demonstrating that it might play a role in the inhibition
